ipaPy2

ipaPy2 implements probabilistic annotation of LC-MS/MS untargeted metabolomics data to provide statistically rigorous metabolite identification using the Integrated Probabilistic Annotation (IPA) framework.

Key Features:

• Python IPA implementation: A Python-based implementation of the Integrated Probabilistic Annotation (IPA) method for probabilistic metabolite annotation.
• Bayesian-based methodology: Employs a Bayesian framework to estimate annotation probabilities and provide statistically principled identification scores.
• Integration of tandem MS fragmentation data: Incorporates MS/MS fragmentation spectra and fragmentation patterns to refine annotation probabilities.
• Isotope peak integration: Aggregates isotope peaks into isotope fingerprints instead of treating them as separate features to improve computational efficiency and annotation consistency.
• Compatibility with mzMatch/PeakMLViewerPy: Integrates with the mzMatch pipeline and PeakMLViewerPy for incorporation into PeakML-based metabolomics workflows.

Scientific Applications:

• Untargeted metabolomics annotation: Provides statistically robust metabolite identifications from LC-MS/MS datasets.
• Systems biology: Enables high-confidence metabolite assignment for metabolic network and pathway analyses.
• Pharmacology: Supports identification of drug-related metabolites and metabolic response profiling.
• Toxicology: Assists in detecting and annotating xenobiotic and endogenous metabolites in toxicity studies.
• Personalized medicine and biomarker discovery: Facilitates discovery and validation of metabolite biomarkers through improved annotation confidence.

Methodology:

Bayesian probabilistic annotation via the Integrated Probabilistic Annotation (IPA) framework implemented in Python; incorporation of MS/MS fragmentation spectra; aggregation of isotope peaks into isotope fingerprints; compatibility with the mzMatch pipeline and PeakMLViewerPy.